Keeping a vehicle well-localized within a prebuilt-map is at the core of any autonomous vehicle navigation system. In this work, we show that both standard SIR sampling and rejection-based optimal sampling are suitable for efficient (10 to 20 ms) real-time pose tracking without feature detection that is using raw point clouds from a 3D LiDAR. Motivated by the large amount of information captured by these sensors, we perform a systematic statistical analysis of how many points are actually required to reach an optimal ratio between efficiency and positioning accuracy. Furthermore, initialization from adverse conditions, e.g., poor GPS signal in urban canyons, we also identify the optimal particle filter settings required to ensure convergence. Our findings include that a decimation factor between 100 and 200 on incoming point clouds provides a large savings in computational cost with a negligible loss in localization accuracy for a VLP-16 scanner. Furthermore, an initial density of ∼2 particles/m 2 is required to achieve 100% convergence success for large-scale (∼100,000 m 2 ), outdoor global localization without any additional hint from GPS or magnetic field sensors. All implementations have been released as open-source software.
Multi-agent system research is a hot topic in different application domains. In robotics, multi-agent robot systems (MRS) can realize complex tasks even if the behavior of each individual agent seems simple thanks to the cooperation between them. Although many control algorithms for MRS are proposed, few experimental results are validated on real data, being essential to building new testbeds to conduct MRS research and teaching. Moreover, most existing platforms for experimentation do not offer an overall solution allowing software and hardware design tools. This paper describes the design and operation of Robotic Park, a new indoor experimental platform for research in multi-agent systems. The heterogeneity and flexibility of its configuration are two of its main features. It supports control strategies design and validation of MRS algorithms. Experiences can be carried out in a virtual environment, in a physical environment, or under a hybrid scheme, as digital twins have been developed in Gazebo and Webots. Currently, two types of aerial vehicles, the Crazyflie 2.X and the DJI Tello, are available. It also includes two types of differential mobile robots, the Turtlebot3 and the Khepera IV. Both internal and external positioning systems using different technologies such as Motion Capture or Ultra-WideBand are also available for experiences. All components are connected through ROS2 (Robot Operating System 2) which enables experiences under a centralized, distributed, or hybrid scheme, and different communication strategies can be implemented. A mixed reality experience that addresses the problem of formation control using event-based control illustrates the platform usage.
<p>Este trabajo presenta el modelado y control completo del sistema Drive-by-Wire de un vehículo urbano eléctrico. Dicho sistema comprende el mecanismo de dirección, la aceleración y el freno del vehículo. El modelado se ha realizado empleando funciones de transferencia de bajo orden haciendo uso de modelos de “caja negra”. En lo referente al control, todos los controladores desarrollados son del tipo PID en sus distintas configuraciones. Los actuadores de corriente continua acoplados a la dirección y freno se controlan mediante un sistema de control en cascada mientras que la aceleración está controlada por un sistema de planificación de ganancias. El código del proyecto se encuentra disponible en la plataforma Github. Los resultados obtenidos demuestran la validez de los modelos obtenidos, así como la eficacia de los controladores desarrollados.</p>
En este trabajo se presenta el diseño de un controlador basado en eventos para el control de altitud del Crazyflie 2.1. La novedad estriba en la implementaci ón del generador de eventos mediante un umbral relativo. La adaptación de este umbral se realiza en función de la distancia a la referencia y la presencia de ruido en el sistema. Los resultados obtenidos muestran una reducción del 93% de actualizaciones de la señal de control requeridas frente a controladores basados en eventos con umbral fijo o controladores de tipo periódico.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.